Open aaronchantrill opened 5 years ago
A simple overview on how to do this would be to organize each news into it's own category, and then note down which ones are being viewed the most. Then, give news accordingly.
(I'm not sure if this has already been implemented as the issue is 4 years old.)
Detailed Description
I'd love to see someone incorporate a real news aggregator into Naomi that would learn your interests and suggest stories based on things it has figured out you are interested in.
That would actually make a good NLTK project, scraping a bunch of feeds, searching for keywords, organizing by relevance, then learning which feeds you click on. Shouldn't be too difficult. I wonder if it could be used to help people understand how filter bubbles work. There's an interesting white paper here: http://www.personal.psu.edu/~gjz5038/paper/www2018_reinforceRec/www2018_reinforceRec.pdf
Context
Possible Implementation
Your Environment